ARCHiv=8
MASSACHOrET - INSTITUTE
OF r'ECrHNOLOLU,Y
FEB 2 5 2015
LIBRARIES
Lifecycle of Viral YouTube Videos
by Ami M. Wang
Submitted to the Department of Architecture and Planning
in partial fulfillment of the requirements for the degree of
Bachelor of Science in Architecture
at the
Massachusetts Institute of Technology
June 2014
@ 2014 Ami M. Wang
All rights reserved
The author hereby grants to MIT permission to reproduce and to distribute
publicly paper and electronic copies of this thesis document in whole or in part in
any medium now known or hereafter created.
Signature of Author:
Department of Architecture
May 23, 2014
S i
n
t
r
red
Signature redacted
Certified by:
Gediminas Urbonas
Associate Professor of Art, Culture and Technology
Thesis Supervisor
Signature redacted
Accepted by:
J. Meejin Yoon
Associate Professor of Architecture
Director of the Undergraduate Architecture Program
-
-1
cte
-2-
Lifecycle of Viral YouTube Videos
by
Ami M. Wang
Submitted to the Department of Architecture on May 23, 2014
in Partial Fulfillment of the Requirements for the Degree of
Bachelor of Science in Architecture
ABSTRACT
YouTube was founded in 2005 as a video-sharing website. Today, it's a powerhouse
social media platform where users can upload, view, comment, and share content. For
many, it's the first site visited when looking for songs, music videos, tv shows, or just
general entertainment. Along with the sharing potential provided by social media like
Twitter, Facebook, Tumblr, and more, YouTube videos have the potential to spread like
wildfire.
A term that has been coined to describe such videos is "viral videos." This comes from
the scientific definition of viral, which involves the contagious nature of the spread of a
virus. Virality on the Internet is not a new concept. Back when email was the hottest
new technology, chain e-mails spreading hoaxes and scams were widely shared by
emailing back and forth. As the Internet aged, however, new forms of virality have
evolved.
This thesis looks at a series of 20 viral videos as case studies and analyzes their growth
over time via the Lifecycle Theory. By analyzing viral videos in this manner, it aids in a
deeper understanding of the human consciousness's affinity for content, the sociology
of online sharing, and the context of today's media culture. This thesis proposes that the
phenomenon of virality supports the claim of Internet as heterotopia.
Thesis Supervisor: Gediminas Urbonas
Title: Associate Professor of Art, Culture and Technology
- 3-
A
Table of Contents
Abstract
3
Part 1: Introduction
7
Part 2: Preliminary Research
9
9
2.1 Concept of Virality
2.2 Content of Viral Videos
11
2.3 Social Spread of Viral Videos
13
2.4 Lifecycle Theory of Viral Videos
15
Part 3: Methodology
16
Part 4: Analysis
17
4.1 Case Study: Growth Phase
18
4.2 Case Study: Peak Phase
20
4.3 Case Study: Decline Phase
21
4.4 Internet as Heterotopia
24
Part 5: Conclusion
26
5.1 Improvements
26
5.2 Future Exploration
26
Bibliography
28
Appendices
29
-5-
-6-
Part 1: Introduction
On February 10th, 2011, a video was uploaded to the video sharing site YouTube
of a young girl singing about her everyday life and dancing with her friends. Completely
unbeknownst to this 13-year-old girl, by March of 2011 she would become a household
name. This video was entitled "Friday - Rebecca Black". After being posted on the
popular blog The Daily What, the video shot to popularity. Most viewers commented on
the video's absurd lyrics, monotone singing, and over-production. After being shared in
high volume, the video had earned 3 million views by March 15th, 2011 and 10 million
views by March 17th, 2012 (Rebecca Black - Friday, 2012). This Internet phenomenon
of the rapid rise of a piece of content to popularity is known as virality.
"Web video has made it so that any of us, or any of the creative things that we
do, can become completely famous and a part of our world's culture" (Allocca, 2011). In
his 2011 TEDtalk, Allocca explains the role of the viral video in online media culture.
Because of their intense rise to popularity and depth of reach, not only can people
upload videos for the whole world to see, but others can express themselves creatively
through blogging about, commenting on, and making parodies of these videos. Because
of this intense participation of audience members, people today now are able to feel a
sense of "ownership in [their] own pop culture" (Allocca, 2011), which is something that
media in the past lacked.
How has this phenomenon emerged and what are the dynamics behind a viral
video? Is there certain type of content that is sure to reach virality? Or is the sharing
aspect more important? Can viral videos be analyzed more deeply by their lifecycles?
This thesis aims to answer these questions to clear up the mystery surrounding viral
-7-
videos to aid in a deeper understanding of the human consciousness's affinity for
content, the sociology of the online network, and the context of today's media culture.
Using a set of viral video case studies, this thesis proposes that the phenomenon of
virality supports the claim of Internet as heterotopia. Additionally, there is a potential
application in the advertising discipline to more profoundly understand a potential niche
market.
- 8-
Part 2: Preliminary Research
2.1 Concept of Virality
According to the Oxford Dictionaries Online, the definition of the term virality is
"the tendency of an image, video, or piece of information to be circulated rapidly and
widely from one Internet user to another." This term has emerged in recent times to
describe this Internet phenomenon - even the definition uses modern terms. However,
the concept of virality is not a new one. The term comes from the scientific definition of
viral, which involves the contagious nature of the spread of a virus. Certain content on
the Internet tends to spread just as infectiously as a viral diseases do, so this term has
arisen in recent years to describe this phenomenon.
However, the concept of virality is not a new one. Communication itself has
changed rapidly throughout mankind as new technologies have been discovered. From
primitive cave drawings to passed down folk tales to newspaper articles, humans have
always found ways to share interesting content with each other.
Chain letters are a phenomenon that took advantage of the growing facilitation of
communication. The earliest chain letter is hard to date, but by the 1900s they were a
common sight (VanArsdale, 2007). Basically, they were letters sent through the post that
asked the receiver to pass the letter along to a few more people. Through this spread,
the number of people reading the letter would begin to increase exponentially. People
would take advantage of this accelerated spread by using chain letters to collect money,
raise awareness for specific causes, encourage prayer, etc.
- 9-
With the invention of the Internet came huge opportunities for rapid
communication. In the early years of the Internet's life, chain e-mails evolved as one of
the first forms of viral content as a virtual form of the chain letters that had already
existed (Mikkelson, 2009). Chain e-mails functioned similarly to their chain letter
counterparts - they consisted of some sort of message that would get the reader to send
the e-mail along to as many people as possible in their e-mail address book. But since
the Internet is infinitely faster than the postal service and much easier to use, chain emails could be spread incredibly quickly.
Mikkelson (2009) broke contemporary chain e-mails into five main categories:
money-generating, luck-generating, altruistic, something for nothing, and humor, and we
can think of these categories in the context of communication in general as well.
Regardless of category, however, most chain e-mails followed the same format. They
began with some sort of story, statistic, quiz, poem, quote, or something else to draw in
the reader. Afterwards, there would be an ask for the user to forward the email to either
a specified number of people or as many people as possible in a certain number of
days. Finally, there is the potential threat or danger for not following instructions,
oftentimes the threat of death or bad luck was used.
By breaking down chain e-mails into this structure, it becomes clear why they
became so viral. First of all, they contained attention grabbers in the form of poems,
stories, or quotes. Next, the act of forwarding to friends is a very low barrier to entry or
challenge for the reader. It's incredibly easy to just press a button and type a few names
to forward an email. Of course, the threat at the end also serves to persuade people to
share. As VanArsdale mentions in his research, the nature of the spread of chain letters
10
-
-
and e-mails can be attributed to population dynamics and distribution networks; the
exponential growth from the few original recipients causes the content to spread in a
viral nature. These chain letters are a prime example of the concept of virality - before
the modern term even existed.
As the Internet has evolved and new applications and websites have been
developed, chain e-mails have waned in popularity. Visual and audio content is now
much more attractive because of the interactivity they offer. Some of the most shared
and most talked about pieces of content are now viral videos stemming from the popular
site YouTube.
2.2 Content of Viral Videos
YouTube was founded in 2005 as a public video-sharing website, available for
anyone to use (About YouTube, 2013). Today, it's a powerhouse social media platform
where users can upload, view, comment, and share content. For many, it's the first site
visited when looking for songs, music videos, tutorials, television shows, or just general
entertainment. Coupled with the sharing potential provided by social media applications
like Twitter, Facebook, Tumbir, and more, YouTube videos have the potential to spread
like wildfire. The platform YouTube created is an essential part of the evolution of the
viral video and of the term virality.
Once this phenomenon started to appear, many individuals began to question
what it takes to create a viral video themselves. There has been much research gone
into the study of the content of viral videos to find common themes: cute animals, irony,
humor, pain, etc. Individuals could take these themes and create their own viral videos.
-
- 11
Big corporations could utilize these tactics for marketing purposes and create viral video
campaigns. West's (2011) research, in particular, uses a series of case studies to
examine the most popular elements of viral videos. The findings from this research is
summarized below.
90%
element of irony
75%
a short title length (three words or less)
60%
a short run-time (three minutes or less)
60%
musical quality
50%
element of surprise
35%
youth
30%
talent
30%
element of laughter
20%
ethnic minority presence
West mentions that many of these qualifications, like irony or talent, are
subjective to the viewer's preferences. That particular fact is what makes it so difficult to
try and predict if a video is going to go viral or not. Even if there were a video that
contained all the elements above, there's little to no way to tell whether or not the
Internet is going to propagate it to viral status or not. The subjectivity of the collective
audience very much affects whether or not a video will rise to popularity, since viral
videos, by definition, are launched to fame mostly through social shares as opposed to
through organic search.
-
- 12
2.3 Social Spread of Viral Videos
Research conducted by a team of engineers at Google in 2010 further examines
this definition of viral videos by separating social and non-social views. The research
analyzes what exactly sets viral videos apart from non-viral videos that still have many
views. In their categorization, they define the "socialness" of videos and how that relates
to the virality of a video.
Broxton and co. analyzed the views and referrals of 1.5 million YouTube videos.
Referrals track the method through which a viewer found the video and can be broken
down into two categories: social and non-social. Social referrals include embedded
videos on blog posts, links shared on Facebook, links sent via instant message, and
other methods that involve peer-to-peer interaction. Non-social referrals include typing
through a search engine or clicking internally within YouTube. By breaking down the
referrals into these two categories, Broxton and co. quantitatively categorized each
video into a specific level of socialness. For example, if 90% of a video's referrals came
from social sources, then it would be considered highly social.
Through their research, they found that the level of socialness of videos varies
significantly among different categories. Figure 1 compares two histograms of videos
from two different categories: "Music" and "Pets & Animals". On the x axis is the
percentage of social views, as defined by the referrals explained above, and on the y
axis is the fraction of videos examined that fall in those buckets. In the "Music" category,
only 0.05 of the videos fall into the 90-100% social category. In the "Pets & Animals"
category, however, more than a fourth of the videos fall in the 90-100% social category.
Viral videos often contain animals, so this is not surprising. While music videos can be
13
-
-
very popular and boast millions of views, the argument is that they would not
necessarily be classified as viral because of their general lack of socialness (Broxton,
2010). There are, however, exceptions - some music videos can be considered viral,
like Gangnam Style or Rebecca Black's Friday.
Category Segmenteton by Percenste of Social Views
Category Segmentaton by Percente of SocIal Views
a
I
Category
Music
Category
I
0
5
'8
C
Pets &Arnmals
CR
d
IL
S.
C
0%
20%
40%
60%
6%
100%
Percentile of Socal Views - First 30 Days
0%
20%
40%
80%
80%
0%
Percentle of Social Views - First 30 Days
Figure 1: Socialness of Music vs. Pets &Animals
Another interesting insight from this research is the difference in level of
socialness of videos over time. They compared two specific cases - a "ViralVideo"
versus a "MusicVideo". As you can see in Figure 2, for the MusicVideo, most of its
referrals are social at the very beginning. But, afterwards, the views are mainly nonsocial and level off to a fairly steady place. In comparison, the ViralVideo relies heavily
on social referrals all throughout its existence, peaking and falling at various intervals
due to the randomness of social sharing. This strong tie to social referrals is what
-14-
separates viral videos from other kinds of videos and is what creates such a strong
impact in the minds of the viewers.
Praciew CSFigu
Vr DY
Vi oses V.ws vs. Day
qq0.
*0
1
0
80
capyso
100
POOPek iew
160
Daso i.Pak View
40
20
-20
0
20
40
60
M010
Figure 2: ViralVideo vs. MusicVideo
2.4 Lifecycle Theory of Viral Videos
What I found to be acutely interesting about viral videos were the various trends
in the ups and downs of their views throughout their lifetimes - how the viewing nature
of a video changes throughout its existence. I became curious about how a viral video
gets started, falls into fame, and then either dies or lives on. Shaw (2013) describes this
phenomenon as the Lifecycle Theory of viral videos. Although the term lifecycle is
usually used in biological contexts, it is extremely relevant here to describe how the viral
video acts. Looking more profoundly into the Lifecycle Theory is the core of this thesis.
-
- 15
Part 3: Methodology
To begin the research for this thesis, first a set of viral videos to examine was
selected. 20 viral videos were selected based on the criteria explained above - videos
that had a rapid rise to fame due to social shares and videos that generally have high
socialness. The 20 videos were complied from various sources listing viral videos from
the past few years and range from 2006 - 2013. The full list of videos selected is shown
in Appendix A.
Next, a method of collecting data for these videos was selected. Because of
limitations on what data is publicly available, I was unable to acquire quantitative video
view data over time. The next best option was to collect Google Trends data for each
video. Although this data, which is web search data, is not exactly the same as video
view data, it is still highly representative of the popularity of the videos over time. The
quantitative data over time collected from Google Trends was extracted and analyzed
for each video.
Additionally, qualitative data was collected for each video through additional
research. Information such as upload date, creator of the video, history behind the
video, and more was gathered. The context behind the history of each video was
essential for making more sense of the quantitative data.
Together, this data was analyzed for trends, outliers, and interesting stories. The
final analysis and a few specific case studies are presented below.
-
- 16
-M=-........
E1111
WE ~~~~~~~i~1
Part 4: Analysis
Figure 3 below shows the data for one video, Harlem Shake, from January 2013
to May 2014 as a representation of the typical shape that these videos exhibit from start
to finish. They begin at a near-zero level, have an extremely rapid rise, and then die
down at a slower rate back to a near-zero level.
Harlem Shake
120
100
80
60
40
20
40
a
C1
NN
OCOi
~O~~
NN
NNN
NN N
0 0
04 N
,9T9o
C C
NN9N
N
.<
N (,
rL7
N
N
NNNN9NN
N
N
N~5
5
N
~
N
N
.
NN
o~
NJ
Fiur 3: Harle Shk dat ove timet
FromR;
that ~
~
~
The~~~~~~
thi comn hpeta i oudaon lms llte iesIprps
~ t th Lieyl Theor be; diieNnotredsic at:GotPaadDcie
~~~
Grwt phs begin whe th vie isN upoddan OatntltePekpae
-17-
NNN O
which is when the video has reached its highest level. The Decline phase starts at the
peak and lasts indefinitely. The following case studies will examine each of these stages
more closely.
4.1 Case Study: Growth Phase
In all 20 of the videos examined, it took between 7-14 days to reach the peak
from a start of zero. Relatively, this is a very short amount of time for something to go
from being completely unknown to being extremely popular. This supports the notion
that viral videos, aided by the sharing power of social media, can rise to fame in an
incredibly short amount of time. How and why does this shot to fame happen? A closer
look at the case study Double Rainbow is quite revealing.
A video called "Yosemitebear Mountain Double Rainbow 1-8-10" was uploaded
on January 8th, 2010 by user Hungrybear9562 and features a man reacting fervently to
a double rainbow in the sky. This particular video, commonly referred to as just Double
Rainbow, caught my attention because of the considerable gap between the upload
date and the start of its rise. In Figure 4 below, the date of the video's upload is
-
highlighted in red. However, the video did not begin to gain popularity until much later
the video sat dormant on the Internet for months before it reached virality.
Why is it, then, that this video suddenly gained popularity? After researching the
history behind this video, the reason became clear. Jimmy Kimmel, a famous talk show
host and comedian, tweeted about the video to his almost three million followers on July
3rd, 2010, declaring that it was "the funniest video in the world" (jimmykimmel, 2010).
Once the video's content reached his followers, they in turn began to share it with their
-
- 18
--..
-- --....
--11
''-mmmmM
friends and families. News networks and other celebrities also began to share the video.
People like Jimmy Kimmel have been dubbed "tastemakers" - people with high social
influence that are often the ones to launch viral videos (Allocca, 2011). Tastemakers can
be celebrities like Jimmy Kimmel or popular blogs like The Daily What. These are
people and places that people trust to tender interesting content and that they want to
be associated with. The sharing is not just about wanting to share a funny video, but
also about playing a part in the creation of something new.
120
-
100
60
40
_
_
_
_
_
_
_
20
4:
Figure
Allocca
participators
asserts
-
not
that
necessarily
in
Double
reality,
the
it's
content
these
over
data
Rainbow
tastemakers,
itself
- 19-
-
that
time
sharers,
truly
perpetuates
and
community
a
video
to
virality.
-
Double Rainbow
While the content is important to create interest, the spread is what sets viral videos
apart from other kinds of videos.
4.2 Case Study: Peak Phase
After a video gains an audience, it creates a vessel for community participation.
The audience of a video can react by commenting on the video, showing it to his or her
brother at the dinner table, or tweeting about it. This creates a community of people who
are now sharing a fun "inside joke" with each other and with a celebrity, which allows for
even more re-shares and re-views. Further participation comes in the form of parodies,
in which viral videos ignite creative energy in others. Parodies are an integral part of the
community of participation, and only act to further the influence of the original video
itself. At this point is usually when the video reaches is Peak Phase, after about 7-14
days. To understand this key concept of participation and parodies, Harlem Shake is a
great case study to examine.
Although the origin of the Harlem Shake phenomenon is not entirely clear, it is
generally accepted that the first video was uploaded by user DizastaMusic on February
2nd, 2013 (Shaw 2013). The video features four men in what looks like a college room
dancing to the trap song "Harlem Shake" by DJ and producer Baauer. Once the beat
drops, the dancing goes from subtle to extremely over-the-top. Other YouTubers
uploaded their own versions of this video that contained the same elements - subtle
dancing that turns extreme and the same song. These parodies started to become more
and more prevalent. The premise was simple, easy to film and edit, and easy to
-20
-
creatively parody. For this reason, the Harlem Shake trend eventually rose to virality.
Millions of shares, comments, likes, parodies, had been produced. Even companies like
Intel and Google produced their own versions of the Harlem Shake within about 2
weeks of the original video's posting.
This level of interactivity allowed Harlem Shake to become more than just a viral
video - it became a global phenomenon. Because YouTube truly allows anyone to
upload content, it creates a platform for everyday people to express their creativity.
People today enjoy creating content and playing a part in creating their own media
culture.
4.3 Case Study: Decline Phase
However, viral videos never stay in the spotlight for too long. After the Peak
Phase, they begin to decline at various rates back down to a near-zero level. The time it
takes to reach this level tends to be around 4-5 weeks for the videos studied in this
research. When compared to something like a Hollywood movie, the entire lifecycle for
a viral video is incredibly short. In just a few short weeks, a video goes from not even
existing, to having billions of views, to being completely irrelevant. This seems to
indicate that for Internet content, people have incredibly short attention spans. Because
of the sheer quantity of content to consume, the life of each one is much shorter. What
does it take for a viral video to extend its lifespan?
A viral video that was able to do just that is the case Nyan Cat. This is an 8-bit
animation of a cat flying through space accompanied by high-pitched, Japanese music.
-
21
-
The trending for this viral video, however, is not the same as the many others, as seen
in Figure 5. Nyan Cat has a much slower decline and still has yet to reach the near-zero
level.
Nyan Cat
120
100 -
-
-
-_
-___
----
---------------------
_______
-__
-
-----------------
---------------
-
60
60
40
-
_--
- ----
-_____________
-----
0
1"
CO
10,C
0c9V~.0
V
-0
7--9
9V
19 j9
t
-r 99
Figure 5: Nyan Cat data over time
Why does this video behave differently? After this video reached virality, it spread
throughout the Internet through various mediums and became what is referred to as a
meme. People created songs, websites, challenges, games, and more based off of the
Nyan Cat video. Because of its adaptation from a simple viral video to a full-fledged
Internet meme, Nyan Cat has been able to live on and significantly extend its lifespan
from the usual 4-5 weeks.
22
-
-
..........
11
...................
..- .1...
..
.......
-- -----
Another interesting case to look because of its Decline Phase is Rickroll. This
video is simply the music video of singer Rick Astley's 1987 single, "Never Gonna Give
You Up". The video, however, shot to popularity through its use as an online prank.
Fellow users can trick each other by pretending to link to something interesting that they
know will be clicked, but instead, it is just Rick Astley's video. This prank began to be
called the Rickroll. A closer look at the data in Figure 6 reveals that after its Growth
Phase and Peak Phase, during every April of its Decline Phase there is a spike in
interest because of friends pranking each other on April Fool's Day. Although Rickroll is
dying down as the years go by, it is an example of a viral video that was able to alter its
lifecycle in an interesting manner.
Rickroll
120
-
-___--______-_
__-
100
80
40
20
A
Figure 6: Rickroll data over time
-23-
4.4 Internet as Heterotopia
After examining these viral videos via this Lifecycle Theory, the dynamics of the
viral video phenomenon are more clear. A trendsetter launches the video to popularity,
and it shoots to its peak through shares, comments, and parodies in around 7-14 days.
Afterwards, the interest in the video slowly declines and 4-5 weeks later, the interest is
back to a near-zero level.
This analysis of virality specifically through viral videos supports the theory of
Internet as heterotopia. Michel Foucault's "Of Other Spaces" brings up the notion of
heterotopic space: socially constructed "other" places that stand outside of known space
(Sherman Young, 1998). Places like cemeteries, zoos, boats, stadiums, or the more
abstract example of a mirror.
Because of the way this research has shown that people act on the Internet
through viral video usage, it clearly supports that the Internet is a form of heterotopic
space. Because the Internet is a space without any materialitiy, it exists only virtually.
Even though it takes up no physical space, it is an extremely powerful medium of
communication - this contradiction is highly indicative of heterotopia.
Additionally, we have the Internet's strange relationship with time. As the
research presented, the average lifecycle of a viral YouTube video is incredibly short.
The accessibility and swiftness of the Internet has created this warped reality of time in
its space. This relationship to time is similar to that of museums and fairs, which are
both considered heterotopias.
What is most interesting is that, like other heterotopias, this Internet one is
completely socially constructed by humans. The way these viral videos spread, become
24
-
-
-
popular, and then die down, happens in a way that can only happen on the Internet
-25
-
and humans have socially constructed it to be that way.
Part 5: Conclusion
This thesis proposes the Lifecycle Theory to analyze viral videos and break them
down into three stages: Growth Phase, Peak Phase, and Decline Phase. By combining
specific quantitative and qualitative data, case studies were examined to more fully
understand the growth dynamics. In conclusion, the data speaks to the unique aspects
of today's new media culture and supports the concept of Internet as heterotopia.
5.1 Improvements
As noted previously, the analysis was conducted using Google Trends data.
Because this is just search data, some of the analysis might have been skewed. Raw
video view data could potentially be more revealing, and could be used to do more
complicated trend analysis. If referral information were also available, the socialness of
videos could also play a part. With this more specific information, more intricate
quantitative analysis could have been calculated.
Additionally, it could have been beneficial to supplement this research with an
actual controlled experiment involving people. Gathering opinions in some form of
experiment could have been interesting, since this is such a subjective subject to be
studying.
5.2 Future Exploration
Through my research, I found that videos produced by marketing companies in
-26
-
the more recent years have risen to virality. It would be interesting to look at these viral
video campaigns from big companies to see which ones were able to successfully
capture this niche market.
Another aspect of this field that I was unable to look into is the global nature of
viral videos. Are there differences or similarities between what videos go viral in various
countries? Do different countries have varying sharing trends or patterns? This would
-27
-
definitely be an interesting distinction to make in future research.
Bibliography
(2013). About YouTube. YouTube. Retrieved from http://www.youtube.com/yt/about/.
Allocca, Kevin. (2011, November). Why videos go viral. TED. [Video file]. Retrieved
from http://www.ted.com/talks/kevin-alloccawhy-videos-goviral.html
Broxton, T., Interian, Y., Vaver J., & Wattenhofer M. (2010). Catching a Viral Video. IEEE
S/ASP. Retrieved from http://static.googleusercontent.com/externalcontent/
untrusteddlcp/research.google.com/en/us/pubs/archive/36697.pdf
jimmykimmel. (2010, July 3). my friend Todd has declared this "funniest video in the
world" - he might very well be right http://bit.ly/75ieRc [Twitter post]. Retrieved
from https://twitter.com/jimmykimmel/status/17665533038
Mikkelson, Barbara (2009). Chain Linked. Snopes. Retrieved from http://
www.snopes.com/luck/chain.asp
(2012). Rebecca Black - Friday. KnowYourMeme. Retrieved from http://
knowyourmeme.com/memes/rebecca-black-friday
Shaw, Keith. (2013). The lifecycle of video viral-ity as demonstrated by the Harlem
Shake. NetworkWorld. Retrieved from http://www.networkworld.com/community/
blog/lifecycle-video-viral-ity-demonstrated-harlem-shake
Sherman Young. (1998) Of cyber spaces: the Internet & heterotopias. M/C: A Journal of
Media and Culture 1(4). http://www.uq.edu.au/mc/9811/hetero.php
VanArsdale, Daniel. (2007). Chain Letter Evolution. Silcom. Retrieved from http://
www.silcom.com/~barnowl/chain-leter/evolution.html
Virality [Def. 1]. (n.d.). Oxford Dictionaries Online. In Oxford Dictionaries. Retrieved from
http://www.oxforddictionaries.com/us/definition/american-english/virality
-28
-
West, Tyler. (2011). Going Viral: Factors That Lead Videos to Become Internet
Phenomena. The Elon Journal of Undergraduate Research in Communications.
Retrieved from http://www.elon.edu/docs/e-web/academics/communications/
research/vol2nol /08West.pdf
. .........
. ........
. ....
Appendices
Appendix A: List of videos examined
2006
http://youtu.be/FzRH3iTQPrk
Evolution of Dance
2006
http://youtu.be/dMH0bHeiRNg
Charlie Bit My Finger
2007
http://youtu.be/_OBIgSz8sSM
Leave Brittany Alone
2007
http://youtu.be/kHmvkRoEowc
Chocolate Rain
2007
http://youtu.be/EwTZ2xpQwpA
Keyboard Cat
2007
http://youtu.be/J---aiyznGQ
Dramatic Chipmunk
2007
http://youtu.be/a1Y73sPHKxw
Rick Roll
2008
http://youtu.be/dQw4w9WgXcQ
David After Dentist
2009
http://youtu.be/txqiwrbYGrs
Double Rainbow
2010
http://youtu.be/OQSNhk5lCTI
Bed Intruder Song
2010
http://youtu.be/hMtZfW2z9dw
Chuck Testa
2011
http://youtu.be/UP1DphOWPs
Nyan Cat
2011
http://youtu.be/QH2-TGUlwu4
Friday - Rebecca Black
2011
http://youtu.be/kfVsfOSbJYO
Talking Twin Babies
2011
http://youtu.be/_JmA2CIUvUY
Gangnam Style
2012
http://youtu.be/9bZkp7q1 9f0
Catch the Ice Dude
2012
http://youtu.be/zd7c5tQCs1 I
Ain't Nobody Got Time for That
2012
http://youtu.be/bFEoMO0pc7k
Harlem Shake
2013
http://youtu.be/8vJiSSAMNWw
What Does the Fox Say
2013
http://youtu.be/jofNRWkoCE
-29
-
Sneezing Panda